An optimization case study for solving a transport robot scheduling problem on quantum-hybrid and quantum-inspired...

Area of Science:

• Computational Optimization
• Quantum Computing Applications
• Operations Research

Background:

• Industrial transport robot scheduling presents complex optimization challenges.
• Evaluating emerging quantum and quantum-inspired computing for practical problems is crucial.
• Classical solvers remain the benchmark for optimization tasks.

Purpose of the Study:

• To benchmark D-Wave's hybrid quantum-classical framework, Fujitsu's digital annealer, and Gurobi's classical solver.
• To assess performance in terms of solution quality and runtime for a real-world transport robot scheduling problem.
• To compare three distinct modeling approaches for the optimization task.

Main Methods:

• Development of three mathematical models for the transport robot scheduling problem.
• Implementation and execution of solvers: D-Wave's hybrid framework, Fujitsu's digital annealer, and Gurobi's classical solver.
• Comparative analysis focusing on solution quality and end-to-end execution time.

Main Results:

• The quantum-inspired digital annealer demonstrated promising performance.
• The hybrid quantum annealer showed potential, with specific opportunities identified.
• Direct comparison highlighted the strengths and weaknesses of each approach against the state-of-the-art classical solver.

Conclusions:

• This study offers valuable insights into applying different optimization strategies to real-world problems.
• Findings aid in evaluating the practical applicability and performance of quantum, quantum-inspired, and classical computing paradigms.
• The research provides a workflow for tackling application-oriented optimization challenges.